{"id":"6bb66df2-9199-4b2d-9ef7-d7e62d12671d","arxiv_id":"2505.04591","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The optimized finite-time environmental quantum Fisher information provides an unambiguous Heisenberg-scaling metric, and two dissipative spin sensors achieve N² scaling, with direct photodetection sufficient in one case.","lead":"This paper gives a new way to rank continuously monitored quantum sensors, counting both the sensing time and the number of qubits as limited resources. It presents two many-body sensor setups that reach the theoretical best scaling in qubit number, and one of them can be read out with ordinary photodetection.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The spin-squeezer leg of the central claim rests on an unproven identity: SM Eq. (S59), which is supposed to reduce the no-jump correlator to C_ZZ(τ1+τ2), is left blank, and the explicit correlator is a leading-order cumulant approximation.","rationale":"The reader correctly identified the restricted derivation of Eq. (10) and the cumulant approximation as the most fragile load-bearing premise. My stress-test deepens this: the supplemental material does not actually complete the derivation for the spin-squeezer—Eq. (S59) is blank, so the central identity is asserted rather than proven. This is a concrete missing step, not merely a matter of restricted validity. The paper's numerical points in Fig. 1b could in principle resolve the concern, but the absence of code or data makes independent verification impossible. I also noticed an apparent factor-of-8 inconsistency between Eq. (13) and Eqs. (14)/(18) (the optimized values seem to be B_max/2 rather than 4B_max times the variance), which further calls for a revision of the quantitative statements; however, that error does not affect the N^2 scaling and is secondary. The high-temperature sensor is exact and its results stand. Because the central claim includes the spin-squeezer as one of the two sensors exhibiting Heisenberg scaling and the photodetection advantage, and because that leg has an unproven identity and an uncontrolled approximation, the paper should be accepted only after the missing derivation is supplied or the numerical data are made available for verification. Thus I recommend CONDITIONAL rather than outright ACCEPT.","tokens_in":21141,"tokens_out":29016,"duration_ms":271407,"concrete_test":"Complete the missing derivation of Eq. (S59): for a dissipation-only Lindbladian with a pure dark state and arbitrary jump operator L, prove (or disprove) that ⟨Z e^{iH_eff†τ1} e^{-iH_effτ2}Z⟩ equals the symmetrized two-time autocorrelator CZZ(τ1+τ2)/2 in the regime used. Then, for the spin squeezer with N=4 and N=8, r=ln(8N), compute IE(T) directly from the two-sided master equation (e.g., via the ancilla-qubit numerical method described in the SM) and compare against Eq. (18). If the direct numerical IE differs from Eq. (18) by more than the claimed agreement, the spin-squeezer central claim fails.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim that the dissipative spin squeezer (Eq. 16) achieves I^opt_E ∝ N^2 T^2 and that direct photodetection saturates the QFI depends on applying Eq. (10), IE(T)=IG(T)-4∫∫CZZ(τ1+τ2), to this system. The supplemental material attempts to derive Eq. (10) for the spin squeezer from the pure-dark-state, dissipation-only case, but the critical step is missing: Eq. (S59), which should show that the two-time no-jump correlator ⟨Z e^{iH_eff†τ1} e^{-iH_effτ2}Z⟩ equals the full symmetrized autocorrelator CZZ(τ1+τ2), is blank in the manuscript. Without this identity, Eq. (10) is not established for the spin squeezer. Furthermore, the explicit form C_JxJx(τ)≈2⟨ΔJ_x^2⟩e^{-2Γτ} used in Eq. (17) comes from a leading-order cumulant expansion and a large-r stationary approximation (SM Eq. S66), not from an exact solution. The paper cites numerical agreement in Fig. 1b, but no code or data are provided to allow independent verification. If either the missing identity or the cumulant approximation fails, the closed-form N^2 scaling and prefactor for the spin squeezer need not hold. The high-temperature superradiant sensor of Eq. (11) is exact and is not affected by this concern; the issue is specifically the spin-squeezer support for the paper's headline result.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper addresses the definition of Heisenberg scaling in continuous quantum sensing. It proposes the optimized finite-time environmental QFI, I_E^opt(T) = max_{Γ_i} I_E({Γ_i};T), as a resource-correct figure of merit, argues that the asymptotic sensitivity S_Z is inadequate because it can diverge with N through slow sensor timescales, and proves a general bound I_E^opt(T) ≤ N^2 T^2 for collective generators, with a tightened filter-function bound 0.262 N^2 T^2 for the restricted class studied. Two sensors are analyzed: a high-temperature superradiant model, for which the environmental QFI is computed exactly and optimized, and a dissipative spin squeezer, for which a leading-order cumulant calculation yields the same N^2 scaling. The paper further claims that for the spin-squeezed sensor, direct photodetection of the cavity output saturates the QFI, avoiding the coherent-absorber decoder needed for the high-temperature sensor.","tokens_in":21544,"tokens_out":8514,"duration_ms":82845,"significance":"The proposed metric directly addresses a known ambiguity in continuous metrology, and the explicit pathological examples in the SM (Kac-factor scaling, uncorrelated-qubit series) make the case concretely. The high-temperature superradiant sensor is solved exactly, and the optimized result I_E^opt ≃ 0.1912 J(J+1)/3 T^2 is a useful benchmark. The bound I_E ≤ 0.262 N^2 T^2 is a clean, parameter-free statement. The dark-state photodetection result is practically significant: if valid, it removes the need for complex decoder networks in a broad class of dissipative sensors. The main unresolved issue is that the spin-squeezer half of the central claim rests on an incomplete derivation and an uncontrolled approximation, so the headline 'striking advantage' is not yet fully established. No computational code or data is provided for the numerical checks.","major_comments":[{"comment":"The identity that reduces the double integral ⟨Z e^{i H_eff† τ1} e^{-i H_eff τ2} Z⟩ to the symmetrized autocorrelator C_ZZ(τ1+τ2) (or an equivalent form) is left blank in Eq. (S59). This identity is the final step in deriving Eq. (S52), i.e. Eq. (10) of the main text, for the dissipative spin squeezer. Without it, the application of Eq. (10) to the spin squeezer, and hence Eq. (18) and the associated photodetection-optimality claim, are not established. Please supply the missing derivation.","section":"Supplemental Material, Eq. (S59)"},{"comment":"The exponential form C_JxJx(τ) ≈ 2⟨ΔJ_x^2⟩ e^{-2Γτ} is obtained from a leading-order cumulant expansion combined with a large-r stationary approximation, with no estimate of the neglected O(e^{-4r}) terms or of higher-order cumulants. The closed-form result Eq. (18) and the claimed N^2 scaling for the spin-squeezed sensor inherit this approximation. The numerical agreement cited in Fig. 1(c) is not independently verifiable because no code or data are included. Please provide either a rigorous error bound on the approximation, a derivation of the correlator beyond leading order, or the numerical data and code used for Fig. 1(c).","section":"Main text, Eqs. (17)-(18); SM Eq. (S66)"}],"minor_comments":[{"comment":"The footnote contains the unresolved placeholder 'Eq. XXX'; it should refer to Eq. (2).","section":"Main text, footnote [32]"},{"comment":"The cross-reference '(see Sec. )' is left blank; please fill in the correct section number.","section":"SM, 'Photodetection is optimal...' section"},{"comment":"The comparison with full numerical simulation is attributed to 'Fig. 1b', but the corresponding simulation data appear in Fig. 1(c); Fig. 1(b) shows the high-temperature I_E(Γ) curve. Please correct the figure reference.","section":"Main text, paragraph after Eq. (17)"},{"comment":"The label 'Inf.-temp. superradiance model' is inconsistent with the 'high-temperature' terminology used throughout the text.","section":"Fig. 2(a)"},{"comment":"The phrase 'completely pure' could be misread as any pure steady state; the SM correctly specifies that the derivation requires a dark state of a dissipation-only Liouvillian. Please make this condition explicit in the main text.","section":"Main text, Eq. (10)"}],"recommendation":"major_revision","confidential_remarks":"To the editor: The manuscript is a good fit for the journal and the metric proposal is valuable. The missing Eq. (S59) appears to be a compilation/LaTeX error rather than a conceptual gap, and the cumulant issue can likely be addressed by supplying the numerical data or an error estimate. If the authors provide these, I would expect the paper to become acceptable. I do not see grounds for rejection, but the current manuscript does not meet the acceptance bar because one of the two headline sensors rests on an incomplete derivation and an uncontrolled approximation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The thing to know: this paper gives continuous sensing a finite-time figure of merit that actually behaves itself. The optimized environmental QFI, with fixed sensing time and optimized couplings, fixes the embarrassing ambiguity where asymptotic sensitivity can show arbitrary N scaling just because timescales diverge. The N^2T^2 bound is clean, and the tighter 0.262 N^2T^2 filter-function bound is a nice bonus. That part is real progress.\n\nThe paper does two other things well. First, the high-temperature superradiant sensor is solved exactly: the autocorrelators, the QFI, and the optimizer are all closed form, and it gives Heisenberg scaling in all three field directions, which the boundary time crystal does not. Second, the spin-squeezed sensor has a genuinely striking practical feature: because the steady state is a pure dark state, direct photodetection saturates the QFI, no coherent-absorber decoder required. If that holds up, it removes the biggest experimental obstacle in this line of work.\n\nNow the soft spots, in proportion. The main formula, Eq. (10), is derived only for the restricted class where the steady state is maximally mixed with a self-adjoint Lindbladian, or a pure dark state with dissipation-only dynamics. That restriction is acknowledged. The high-temperature leg is safe. The spin-squeezer leg is where I worry. The SM derives Eq. (10) for the dark-state case, but the critical step that reduces the two-time no-jump correlator to C_ZZ(τ1+τ2) is literally a blank equation: Eq. (S59) is missing. That is not a cosmetic omission; it is the load-bearing identity for this sensor. On top of that, the explicit correlator C_JxJx(τ) comes from a leading-order cumulant expansion plus a large-r approximation, not an exact solution. The paper cites numerical agreement in Fig. 1, but no code or data are provided, so an independent check is not possible. These issues do not touch the high-temperature results or the general bound, but they mean the spin-squeezer's closed-form N^2 prefactor is not yet fully supported.\n\nIs it still worth engaging? Yes. The metric is well-motivated, the high-temperature sensor is exact, and the photodetection-for-dark-states result is important even if the spin-squeezer numbers need tightening. This is exactly the kind of paper I would send to a serious referee: the core idea is good, the flaws are localized and fixable, and the missing equation needs to be supplied before publication.\n\nFor you specifically: worth reading for the metric and the high-temperature solution. I would cite the finite-time QFI metric in my own work, but I would wait until the spin-squeezer derivation is complete before citing the squeezing result.","headline":"Useful new finite-time metric for continuous metrology, with an exact N^2T^2 bound and two Heisenberg-limited sensors; the spin-squeezer leg has a real gap in the SM that needs fixing, but the core idea is sound.","tokens_in":21988,"tokens_out":1307,"would_cite":true,"duration_ms":14237,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proposes a finite-time metric for continuous quantum sensing, proves its N²T² bound, and gives two dissipative sensors that meet it.","keywords":["continuous quantum metrology","quantum Fisher information","Heisenberg scaling","open quantum systems","spin squeezing","superradiance","photodetection"],"falsifier":"Measure the estimation error of the dissipative spin squeezer with even $N$, large squeezing $r$, and $\\Gamma\\simeq 1.89/T$ over a fixed time $T$; the claim predicts an error scaling as the inverse of $0.191\\,J(J+1)\\,T^2/2$ with $J=N/2$, so a visibly weaker-than-quadratic $N$ scaling, or a much larger constant, would falsify it.","tokens_in":20939,"feed_emoji":"⏱️","tokens_out":11075,"duration_ms":98584,"temperature":0.7,"pith_summary":"The paper argues that the usual asymptotic 'sensitivity' metric for continuous quantum sensors is misleading, since it can be inflated arbitrarily by making a sensor slow, and proposes instead the optimized finite-time environmental quantum Fisher information, $I_E^{\\mathrm{opt}}(T)=\\max_{\\{\\Gamma_j\\}} I_E(\\{\\Gamma_i\\};T)$, which counts both integration time and system size as finite resources. For a restricted class of dissipative sensors the paper derives an exact formula for this quantity from the stationary noise spectrum, and proves the bound $I_E^{\\mathrm{opt}}(T)\\le N^2T^2$. It then exhibits two $N$-qubit sensors—high-temperature superradiance and dissipative spin squeezing—that reach $N^2T^2$ Heisenberg scaling for collective magnetic fields, with the spin-squeezed sensor having the additional property that its quantum limit is reached by plain photodetection of the cavity output. If the claims hold, experimenters gain a defensible definition of Heisenberg scaling and two concrete setups that achieve it with simple measurements.","feed_headline":"Finite-time metric pins down Heisenberg scaling in continuous sensing","feed_subtitle":"A meter that counts time as a resource separates real Heisenberg gains from slow-sensor artifacts.","key_machinery":"The load-bearing object is the two-sided (pseudo-density) master equation of Eq. (3), whose trace yields the global QFI and whose modulus yields the environmental QFI. For the restricted class, Eq. (10) converts these QFIs into integrals of the stationary symmetrized autocorrelation function $C_{ZZ}(\\tau)$, with the environment contribution being the global contribution minus a correction built from the two-time correlation $C_{ZZ}(\\tau_1+\\tau_2)$. The optimized metric $I_E^{\\mathrm{opt}}(T)$ then captures a tradeoff: very weak waveguide coupling emits no signal, while very strong coupling dephases the sensor before the parameter is imprinted; the optimum sits near $\\Gamma\\sim 1/T$. For the spin squeezer, the second load-bearing object is its pure dark state—a state annihilated by all jump operators—which makes direct photodetection of the output field the QFI-saturating measurement.","core_discovery":"On its own terms, the central discovery is that continuous sensing has a well-defined Heisenberg limit once time is treated as a finite resource. The proposed figure of merit is the optimized finite-time environmental QFI, $I_E^{\\mathrm{opt}}(T)\\equiv\\max_{\\{\\Gamma_j\\}}I_E(\\{\\Gamma_i\\};T)$, which is bounded by $N^2T^2$ for any collective generator $\\hat Z=\\hat J_{\\vec r}$. For sensors with $\\hat H_0=0$ whose steady state is either maximally mixed with Hermitian jump operators or a pure dark state with purely dissipative dynamics, the paper derives the exact identity $I_E(T)=I_G(T)-4\\int_0^T\\!d\\tau_1\\int_0^{\\tau_1}\\!d\\tau_2\\,C_{ZZ}(\\tau_1+\\tau_2)$. Using this formula, the high-temperature superradiant sensor gives $I_E^{\\mathrm{opt}}\\simeq 0.1912\\,J(J+1)\\,T^2/3$ for all three field directions, and the dissipative spin squeezer gives $I_E^{\\mathrm{opt}}[\\hat J_x]\\simeq 0.191\\,J(J+1)\\,T^2/2$, both of order $N^2T^2$; for the spin squeezer the optimal measurement is direct photodetection.","pith_inferences":["A testable extension is to run the same fixed-$T$ optimization on the odd-$N$ spin squeezer, where the paper's numerics suggest $N^2$ scaling without spin squeezing; confirming that would show that environment–system entanglement alone can produce Heisenberg scaling.","If Eq. (10) holds beyond the two proven cases, the environmental QFI of a candidate sensor could be predicted directly from its measured noise spectrum, turning the metric into a practical screening tool.","The same resource accounting could be exported to waveform estimation or finite-bandwidth sensing, where the filter-function form of $I_E$ connects the optimal integration time to the bandwidth of the signal.","A curious open question is whether the exact identity is the leading term of a general relation between environmental QFI and noise spectra, with the maximally-mixed and pure-dark cases as endpoints."],"forward_implications":["Heisenberg scaling in continuous sensing becomes precise: for any collective generator $\\hat Z=\\hat J_{\\vec r}$ and fixed $T$, $I_E^{\\mathrm{opt}}(T)\\le N^2T^2$, so an $N^2$ scaling cannot be an artifact of a slow sensor.","Optimizing the qubit–waveguide coupling of the high-temperature superradiant sensor at $\\Gamma_\\alpha\\simeq 1.89/T$ yields $I_E^{\\mathrm{opt}}\\simeq 0.1912\\,J(J+1)\\,T^2/3$ for fields along $x$, $y$, or $z$.","The dissipative spin squeezer, at large squeezing $r$ and $\\Gamma\\simeq 1.89/T$, yields $I_E^{\\mathrm{opt}}[\\hat J_x]\\simeq 0.191\\,J(J+1)\\,T^2/2$, a prefactor $3/2$ larger than the thermal sensor.","Any sensor hosting a pure dark state saturates its environmental QFI with direct photodetection of the emitted field; no coherent-absorber decoder is required.","The conventional sensitivity $S_Z$ is not bounded by $N$ and can display spurious super-Heisenberg scaling, so it is not a reliable certification of Heisenberg scaling."],"supporting_citations":[{"why":"Supplies the environmental-QFI expression from the pseudo-density matrix and the coherent-absorber decoder construction.","marker":"[9]"},{"why":"Gives the global QFI as an integral of the symmetrized stationary autocorrelation function $C_{ZZ}(\\tau)$.","marker":"[24]"},{"why":"Introduces the two-sided master equation used to compute both global and environmental QFIs.","marker":"[29]"},{"why":"Contains the derivations of Eq. (10), the $N^2T^2$ bounds, the cumulant expansion for the spin squeezer, and the photodetection optimality proof.","marker":"[31]"},{"why":"Provides the boundary-time-crystal sensor whose sensitivity scaling motivates the new metric and whose Heisenberg scaling is only along one axis.","marker":"[33]"},{"why":"Establishes the coherent quantum absorber construction used for optimal measurements.","marker":"[35]"},{"why":"Shows hidden time-reversal symmetry, the criterion that makes simple absorber construction possible.","marker":"[36]"},{"why":"Provides the dissipative spin-squeezer model, its pure dark steady state for even $N$, and the steady-state variances used in Eq. (17).","marker":"[45]"},{"why":"Shows how the squeezed-light pumping plus cavity damping can realize the dissipative spin-squeezer dynamics.","marker":"[47]"}],"fun_headline_variants":["Finite-time metric defines Heisenberg scaling in continuous sensing","Direct photodetection reaches quantum limit in many-body sensing","Heisenberg scaling for continuous sensing with finite resources","Time and size as resources: new metric for Heisenberg scaling","Spin squeezer senses Heisenberg limit via direct photodetection"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is the derivation of Eq. (10), which is proven only for a maximally mixed steady state with a self-adjoint dissipator or for a pure dark state with purely dissipative dynamics, and for the spin squeezer it also rests on a leading-order cumulant expansion and a large-$r$ approximation.","fun_headline_variants_meta":{"raw":{"variants":["Finite-time metric defines Heisenberg scaling in continuous sensing","Direct photodetection reaches quantum limit in many-body sensing","Heisenberg scaling for continuous sensing with finite resources","Time and size as resources: new metric for Heisenberg scaling","Spin squeezer senses Heisenberg limit via direct photodetection"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000644,"raw_usage":{"total_tokens":3004,"prompt_tokens":1029,"completion_tokens":1975,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":645,"completion_tokens_details":{"reasoning_tokens":1894}},"tokens_in":645,"tokens_out":1975,"duration_ms":12690,"temperature":1.0,"reasoning_tokens":1894,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T23:25:22.846488+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the estimation error of the dissipative spin squeezer with even $N$, large squeezing $r$, and $\\Gamma\\simeq 1.89/T$ over a fixed time $T$; the claim predicts an error scaling as the inverse of $0.191\\,J(J+1)\\,T^2/2$ with $J=N/2$, so a visibly weaker-than-quadratic $N$ scaling, or a much larger constant, would falsify it.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the environmental-QFI expression from the pseudo-density matrix and the coherent-absorber decoder construction."},{"cited_title":"Gammelmark and K","cited_arxiv_id":null,"evidence_quote":"Gives the global QFI as an integral of the symmetrized stationary autocorrelation function $C_{ZZ}(\\tau)$."},{"cited_title":"Gammelmark and K","cited_arxiv_id":null,"evidence_quote":"Introduces the two-sided master equation used to compute both global and environmental QFIs."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Contains the derivations of Eq. (10), the $N^2T^2$ bounds, the cumulant expansion for the spin squeezer, and the photodetection optimality proof."},{"cited_title":"Stannigel, P","cited_arxiv_id":null,"evidence_quote":"Establishes the coherent quantum absorber construction used for optimal measurements."},{"cited_title":"Roberts, A","cited_arxiv_id":null,"evidence_quote":"Shows hidden time-reversal symmetry, the criterion that makes simple absorber construction possible."},{"cited_title":"Groszkowski, M","cited_arxiv_id":null,"evidence_quote":"Provides the dissipative spin-squeezer model, its pure dark steady state for even $N$, and the steady-state variances used in Eq. (17)."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows how the squeezed-light pumping plus cavity damping can realize the dissipative spin-squeezer dynamics."}],"review_version":1}